Friday, 14 October 2011

government

I remember Gimson, probably sometime in the 70s, telling me that Jack Windsor Lewis was trying to convince him that he ought to change the EPD entry for government so as to prioritize the variant in which there is no n before the m. Gim didn’t think much of this idea, and continued to prioritize the pronunciation ˈɡʌvnmənt.

In his CPD (1972) Jack transcribes this word as ɡʌvm̩ənt, and entirely iɡnores the possibility of n before the m. See also his comments in §10 of this review.
the term government … can be heard every day over and over again in countless news bulletins and current affairs programmes. Both EPD15 and LPD list first the variant which contains the /-nm-/ sequence. However, anyone who listens at all attentively to recordings will soon discover that this is not merely not the predominantly heard form of the word, even in situations of the greatest prominence or highlighting, but that it is actually even a relatively unusual form of it.

Now Giovanbattista Fichera writes to take me to task over the same issue, expressing surprise that I have not acted on Jack’s criticism. (In LPD the main entry (BrE) for this word continues to read ˈɡʌv ən mənt.) GF continues
In my opinion, it requires a great deal of effort to articulate the <-nm-> sequence without assimilating the n to the m. In Italian, San Mauro is sa'm:auro not san'mauro. It seems to me that mm is not a variant/change in progress, but the rule.

I replied
I think my entry is correct. It unquestionably corresponds to my own slow-careful pronunciation of the word. I certainly don't have to make “a great deal of effort” to pronounce it as shown. (But then my L1 is English, not Italian or Japanese.) The alternatives that follow represent reductions which are also admittedly very common in speech - but they are just that, reductions.


I am in no doubt that for me (at least) ˈɡʌvənmənt accurately represents the succession of articulatory targets presumably stored in my mental lexicon as the phonological specification of this word. As I told GF, it also represents the way I pronounce it when articulating carefully (not overarticulating, but also not applying running-speech reductions).

Some of these reductions may indeed be frequently heard from speakers, in radio or TV news bulletins as elsewhere. That does not make them the mentally stored forms, which are what I think a dictionary ought primarily to record.

As for GF’s comments on San Mauro, he is perfectly correct as far as Italian is concerned. Italian, like Spanish, Japanese and various other languages, does not admit sequences of nasals at different places of articulation. But English does. Even JWL’s CPD shows -nm- at inmate, with no other variant given. (In LPD, on the other hand, I do recognize the possibility of ˈɪmmeɪt, derived automatically by applying the option of dealveolar assimilation.)

41 comments:

  1. John, the pronunciation with [-nm-] in "San Mauro" in Italian IS possible. I sometimes use it, although I have to admit that it's more common in careful speech.
    As you correctly pointed out in your post, native speakers usually assimilate the dental nasal to the bilabial one, thus producing [-mm-].

    ReplyDelete
  2. Like Italians, I assimilate to gʌvəmmnt. That's in careful speech — more often I say gʌvmnt.

    ReplyDelete
  3. The 'gʌvənmnt~'gʌvəmmnt pronunciations are available to me, but it definitely feels like a spelling pronunciation. I think the most natural form is 'gʌvəmnt, with a single -m-.

    So I suppose I'm agreeing with David Crosbie, who doubles the -m- in careful speech.

    Best thing to do would be to watch the news on various channels for a few days and observe how it's pronounced in real life.

    ReplyDelete
  4. John,
    «the way I pronounce it when articulating carefully (not overarticulating, but also not applying running-speech reductions).»

    I absolutely agree in doing that, and with the rest of your listing for 'government', with all the alternatives in that order.

    But although as a hearer I say I agree with the ordering of the subsequent alternatives, as a speaker I'm as sure as it's possible to be in the uncertain world of phonetics that I never use any of them.

    For me what would come second in the order of priority would be ˈgʌɱnmənt, and ˈgʌɱm̩ənt would come immediately after it.

    Obviously you have to be selective with the pretty open-ended list of alternatives for practically everything, but did you make a conscious decision that those forms were not worth mentioning?

    On the other hand I'm surprised you didn't slap a § on some of the ones you did mention! You could have mentioned §ˈgʌvm(ə)nt too if you had. And what would have merited the § would have been the fact that it's so marked that it it's capable of determining a different sign identity, as in 'governor' and 'guvnor, guv’nor', the separate entries for which you of course distinguish by the presence or absence of the raised ə.

    ReplyDelete
  5. what about the word "environment"? Is the same thing applied to it, John?

    ReplyDelete
  6. It might be an illusion caused by the word's spelling ("spelling hearing"?), but I've always thought the Americans actually have /ɡʌvərmənt/, so pronounce it [ɡʌvɚmənt] or [ɡʌvɚmn̩t] or the like. Am I mistaken?

    ReplyDelete
  7. Haven't you got LPD3. ismael? John shows the very close parallels in his entry for that:

    environment ɪn ˈvaɪ ᵊ r ə n mənt en-, ən-, →-əm-, -ə-

    But for some reason although he usually shows the smoothing and compressing options with linking r, as in firearm ˈfa ɪ ‿ər ɑːm , he doesn’t seem to show them where the r is intervocalic or followed by a syllabic consonant, as in ɪn ˈvaɪ ‿ər ən mənt, ɪn ˈvaɪ ‿ər n̩ mənt, ɪɱ ˈvaɪ ‿ər m̩ mənt etc, which would certainly be possible for me.

    ReplyDelete
  8. No I don’t think you are mistaken Kilian. I too am surprised the only form indicated for the US pronunciation is ˈɡʌv ᵊrn mənt . I would have expected to see something like ˈɡʌv ᵊr mənt there too.

    ReplyDelete
  9. The reason why LPD does not record /ˈɡʌv ᵊr mənt/ is probably because it's not one of John's "mentally stored forms". ;)

    ReplyDelete
  10. I suggest that you do a poll on this word for the next collection. I'm confident that you'll find that only a minority of people pronounce the first n. How many politicians say the n in it? Jack Windsor Lewis claimed here in 1990 that the only ones who have consistently said it were Thatcher, Benn and Powell. There is a problem here that Thatcher and Benn speak old-fashioned forms of RP, and Powell wasn't RP at all.

    I think that your LPD polls are good as they introduce some consensus into your model of RP. I get irritated by people who are convinced that their own pronunciation determines what is "received" just because they have the right background.

    ReplyDelete
  11. I do say /ɡʌvɚmənt/, even when speaking lento, and consider it the normal form. /ɡʌvɚnmənt/ would be the rare abstract noun meaning 'act of governing', as in the line from Messiah: "And the government shall be upon his shoulder", or this sentence from The Lord of the Rings: "They used therefore the Common Speech in their dealing with other folk and in the government of their wide realms; but they enlarged the language and enriched it with many words drawn from the Elven-tongues."

    ReplyDelete
  12. John Cowan's US view tallies with my Irish view. For /ɡʌvɚmənt/ "the civic power" vs /ɡʌvɚnmənt/ "process of governing", cf. 'business' vs 'busyness'.

    ReplyDelete
  13. I always hear NM in Italian pronunciation of San Marino, and in Spanish pronunciation of inmigrante, inmunología...

    ReplyDelete
  14. I sometimes hear something like 'gumment', 'gummunt' (the last vowel being a 'shwa').

    ReplyDelete
  15. Well, American English is essentially Hiberno-English with a German accent, yes.

    ReplyDelete
  16. Some thoughts on [nm]: Wolfram Alpha lists 197 words spelt with the sequence -nm-. Excluding a handful of proper names, the words are all morphologically complex with -nm- occurring at a boundary: e.g., govern+ment, adorn+ment, corn+meal, un+made, and so on. The one exception is the word 'enmity'. From this we can conclude that [nm] is actually uncommon within English morphemes.

    The fact that there exists a stem, govern, pronounced with [n], allows us to posit the 'n' in a mentally stored representation. This 'n' is retained in the inflectional form, govern+ing, but it is (perhaps) variously deleted or assimilated in the derived form, govern+ment, depending upon the speaker and dialect.

    Nevertheless, I wonder if there is something to JWL's claim. Since 'government' is such a frequently used word (according to Google Ngram viewer it has always been more frequent that 'govern' and 'governing'; and, according to WA, it is the most common word with -nm-), there is a possibility that some speakers are reanalysing the derived form as a simple form with the n deleted to avoid [nm] in an underlying mental representation, e.g. JWL's 'govment'. If the original derived form is also retained as an option, then this could be correlated with the meaning distinction indicated by John Cowan.

    ReplyDelete
  17. Correction: the example I gave in the third paragraph is wrong since it creates the sequence [vb]. A simple form would have to be something like 'gove(r)ment'. On reflection, I think this is better treated as optional deletion of 'n'.

    ReplyDelete
  18. Pianoman said...

    I always hear NM in Italian pronunciation of San Marino

    Canepari marks this pronunciation with a down arrow, i.e. trascurata ('slovenly' according to my dictionary!).

    ReplyDelete
  19. Hard to believe, isn't it, Steve?

    Now it doesn’t surprise me that Alex says the pronunciation with [-nm-] in "San Mauro" in Italian IS possible… in careful speech, i.e. no "great deal of effort to articulate", as Giovanbattista Fichera claims. But trascurata?

    Could the arrow in Canepari be a misprint? I didn't think he was quite that perverse!

    Like JW, I would have been confident that Giovanbattista Fichera’s comments on San Mauro were perfectly correct as far as Italian is concerned, and that [mm] is not a variant/change in progress, but the rule (even if with Alex's exceptions), though I do think Spanish is a different matter, as in pianoman's examples 'inmigrante' and 'inmunología'. So I would not, as JW does, include Spanish as a language which like Japanese [never] and various other languages [to varying degrees] does not admit sequences of nasals at different places of articulation.

    ReplyDelete
  20. @Pete
    «The 'gʌvənmnt~'gʌvəmmnt pronunciations are available to me, but it definitely feels like a spelling pronunciation. I think the most natural form is 'gʌvəmnt, with a single -m-.»

    So no rhoticism? Are you not the Ulster Pete?

    (If so I hope you read my repudiation of the extraordinary ideas about phonotactic rules you accused me of re Dugher etc, and preferably also reread my earlier comments so as to make sense of them!)

    ReplyDelete
  21. @John C
    «I do say /ɡʌvɚmənt/, even when speaking lento, and consider it the normal form. /ɡʌvɚnmənt/ would be the rare abstract noun meaning 'act of governing', as in the line from Messiah: "And the government shall be upon his shoulder"»

    Spot on! Kilian and I expressed surprise that the only form indicated in LPD for the US pronunciation is ˈɡʌv ᵊrn mənt . I said I would have expected to see something like ˈɡʌv ᵊr mənt there too, but it was you who had to point out the distinction.

    But it does seem I can do predictive, as in my earlier statement that a variant like that can be capable of determining a different sign identity, as in 'governor' and 'guvnor, guv’nor', the separate entries for which LPD of course distinguishes by the presence or absence of the raised ə. Thus a purely realizational matter like these elisions can become subject to conventions which enable them to carry the burden of semantically distinctive function. But they don't necessarily do so on a one-to-one basis. That's actually more unusual than is generally acknowledged.

    So I don't doubt that for you and mollymooly, and indeed most speakers of Hiberno-English with or without a German accent, /ɡʌvɚmənt/ is the *normal* form and /ɡʌvɚnmənt/ would be the [not particularly] rare abstract noun meaning 'act of governing'. But surely you can't be arguing that just because they're *potentially* distinct, they're mutually exclusive forms . There's bound to be overlap between the set of variants of "Government" and the set of variants of "government". Insofar as we can agree that the two are separate lexemes rather than allosemes (as OED makes them, but its assignment of lemmata is erratic), the overlapping variants are homomorphs.

    And that overlap may be proper inclusion: for the (undifferentiated) 'government' in BrE which JW is dealing with here, I agreed with him that ˈɡʌvənmənt (or at least ˈɡʌvn̩mənt) is "the way I pronounce it when articulating carefully (not overarticulating, but also not applying running-speech reductions)."

    But your observation still holds: even if that form with minimal assimilation and elision is not distinctive with respect to my (and no doubt his) other forms with more assimilation and/or elision, those other forms *are* distinctive with respect to that form. The variants of your abstract noun 'government' are for me distinctive with respect to the "Government" we allegedly elect. But not vice versa. That is all it takes to establish two different sign identities.

    @PeterT
    How do you use Wolfram Alpha for that? It doesn’t seem to recognize the asterisk.

    ReplyDelete
  22. @ mallamb and Steve:

    Mr Canepari is so prescriptive he just doesn't stand the sequence [-nm-], but I can assure you it IS possible in Italian. For me (at least), the assimilated form is the 'careless' one and the unassimilated variant the default one.

    If you check on my blog, you'll find lots of other examples which show that Canepari's DIPI just doesn't represent what Italians really say!

    http://alex-ateachersthoughts.blogspot.com

    ReplyDelete
  23. @mallamb

    Just query 'words with nm'.

    ReplyDelete
  24. Suppose that the the following were authentic English words, which could just conceivably come about...

    botherment, motherment, brotherment, coverment, ditherment, shiverment, fatherment, featherment

    ... How many of us would — in careful speech — have the same rhythm is government and coverment? My feeling is that I don't — in careful speech — say gʌvəmnt because I'm subliminally aware of the morphology: i.e. there's no such thing as (a) gʌvə.

    It's not the spelling that triggers -mm-, it's the morphology.

    ReplyDelete
  25. I may have missed something, but it seems that nobody here uses the old Gimson choice ˈɡʌvnmənt.

    In careful speech
    • Those of us who retain v and/or retain n have two syllable before -ment
    • Those of us who have one syllable before -ment do not attempt to articulate n

    We can, it seems, articulate -nm- with relative ease if we slow down. What gives us problems is -vn-. Yes, of course we can say hæv nɔt, and even gʌvnə, but we don't find it natural to sing Heav'n in nineteenth century hymns.

    ReplyDelete
  26. David C: for Gimson the notation ˈɡʌvnmənt implied syllabic . He pronounced the word the same way as I do, with three syllables. I haven't yet seen the brand-new 18th edition of EPD, but the 17th continues to give ˈɡʌv ən mənt (first schwa superscript), just as does LPD.

    ReplyDelete
  27. Thanks for the explanation, John.

    On another tack ...
    Once I started thinking about morphology, it gradually occurred to me that the morpheme {govern} has — for me — lost semantic value in the word government. In its normal senses, that is.

    For me the default meaning of government is 'executive' or such extensions as 'personnel of the executive', 'operation of the executive', 'nature of the executive', 'existence of an executive'. (The last three are increasingly replaced by governance.) By further extension, I think of the intransitive verb govern as 'function as an executive'.

    But there's also a transitive verb meaning 'rule over'. And this can characterise the morpheme in government, for example in the technical syntactic term government and binding. Now I've never once uttered this term aloud. But I have read it, often silently pronounced in my mind's ear. And what I silently say is gʌvənmnt.

    Psychologically, gʌvənmnt ('sway over') and gʌvəmmnt ('executive') have — for me — become different words.

    ReplyDelete
  28. David,
    «I may have missed something, but it seems that nobody here uses the old Gimson choice ˈɡʌvnmənt

    You did miss something. I said I pronounced it ˈɡʌvənmənt (or at least ˈɡʌvn̩mənt) when articulating carefully, and as John says, for Gimson the notation ˈɡʌvnmənt implied syllabic n̩.

    «What gives us problems is -vn-. Yes, of course we can say hæv nɔt, and even gʌvnə, but we don't find it natural to sing Heav'n in nineteenth century hymns.»

    So you missed something else:

    Not only did I say of the LPD main entry for "the (undifferentiated) 'government' in BrE" (ˈɡʌv ᵊn mənt, with John's graphic to prove it), that this represents the way I too "pronounce it when articulating carefully (not overarticulating, but also not applying running-speech reductions)," and that I agreed with the rest of his listing for 'government', with all the alternatives in the order they appeared in (ˈɡʌv ᵊn mənt already implies the syllabic n̩, as with Gimson's ˈɡʌvnmənt), but I also regretted the non-appearance after that main entry ˈɡʌv ᵊn mənt of what are also common reductions, at least for me:

    «For me what would come second in the order of priority would be ˈgʌɱnmənt, and ˈgʌɱm̩ənt would come immediately after it.»

    I'm afraid this pre-empts your claim that "we don't find it natural to sing Heav'n in nineteenth century hymns", and makes it obvious why I don't agree with it. Certainly I quite happily sing heɱn, even as one syllable, and so I think do most people. One hardly ever seems to hear any frication there.

    You then missed a whole lot more. In your post on the morphology and semantic value you say that the potential distinction on that basis has only just occurred to you, but it had been pointed out by John Cowan and analyzed at I think for once necessary length by me.

    If you were being "Honest" in your claim to read previous posts, you don't seem to be reading them with much attention.

    Sorry, got brainfag, but I reposted that after an effort to make it less impenetrable.

    ReplyDelete
  29. John W,
    «I haven't yet seen the brand-new 18th edition of EPD, but the 17th continues to give ˈɡʌv ən mənt (first schwa superscript), just as does LPD.»

    I know you usually do the superscript schwa with the <sup> tag, and that one can't do that in comments, but as you can see from my previous posts you can use ᵊ (U+1D4A) for it, from a dedicated range for a whole lot of other superscripts. I have made AutoCorrect entries for some of them.

    ReplyDelete
  30. I've also put ᵊ on the ALTGR+SHIFT for E on Mark Huckvale's Unicode Phonetic Keyboard. You would be able to tell me how I did that, and to suggest how some of the other unassigned options could be implemented in the program for things like this.

    ReplyDelete
  31. Mallamb

    You did miss something. I said I pronounced it ˈɡʌvənmənt (or at least ˈɡʌvn̩mənt) when articulating carefully, and as John says, for Gimson the notation ˈɡʌvnmənt implied syllabic n̩.

    No I didn't miss that, Mallamb. I assumed, wrongly as John informs me, that Gimson recognised a monosyllabic gʌvn. I included you in the category of those who pronounced n and (therefore?) had two syllable before ment.

    you say that the potential distinction on that basis has only just occurred to you, but it had been pointed out by John Cowan and analyzed at I think for once necessary length by me

    To this one I plead guilty. I didn't read John C's post carefully enough.

    ReplyDelete
  32. David,
    «I included you in the category of those who pronounced n and (therefore?) had two syllable before ment.»

    OK, and I do have two syllables before -ment. Thank goodness we sorted that out!

    But oh dear, of "it had been pointed out by John Cowan and analyzed at I think for once necessary length by me" you say

    «To this one I plead guilty. I didn't read John C's post carefully enough.»

    Not my necessarily long one at all, then.

    ReplyDelete
  33. Mallamb

    Not my necessarily long one at all, then.

    I was taught phonetics by some very good people — but not for long and each time starting from scratch. What I've read since then sometimes gives me the confidence to advance a view, but often I feel out of my depth. That's just how I felt when I read:

    So I don't doubt that for you and mollymooly, and indeed most speakers of Hiberno-English with or without a German accent, ...

    Mentally, I fled.

    ReplyDelete
  34. David Crosbie: When I use the word essentially, it means what I want it to mean, neither more nor less.

    M.A.L. Lamb: Note that AmE does not have the word guvnor in any form (nor does it use governor with the relevant semantics) and when I try to say it with my own accent, I stumble over the /vn/, further evidence that you are right.

    ReplyDelete
  35. David,
    I appreciate your pleasantry, but you don't seem to appreciate it was a pleasantry about a pleasantry of mine, based on some other posts you had apparently not read, namely mollymooly's initial response to John Cowan's morpho-semantic scoop on you:

    «John Cowan's US view tallies with my Irish view. For /ɡʌvɚmənt/ "the civic power" vs /ɡʌvɚnmənt/ "process of governing", cf. 'business' vs 'busyness'.»

    (The point being I suppose that though 'business' vs 'busyness' is of course fully lexicalized, with the sort of allosemes I was talking about being obsolete, and assigned to a lemma with the respelling, unlike the case of 'government', it has gone through the same process as we and you have commented on. But English orthography sees no need to do anything about left-over anomalies resulting from 'business' having also completed the transition to two syllables, again unlike the case of 'government'. I don't think your observations will bring about spellings like Goverment or Gov'ment, Govment any time soon!)

    And John Cowan's crack about mollymooly's Irish view:

    «Well, American English is essentially Hiberno-English with a German accent, yes.»

    So now you see how my little joke fitted in:

    « Hiberno-English with or without a German accent, ...»

    simply means either American English or Hiberno-English.

    ReplyDelete
  36. John Cowan

    When I use the word essentially, it means what I want it to mean, neither more nor less.

    Yes but when you use the terms Hiberno-English and German accent it's not that i don't' know what you mean, but I just don't feel sufficiently well-informed to follow you. The fact that it seems to be a joke (or half-joke) makes it harder, not easier.

    If I understood the rest of the sentence, essentially would be transparent.

    I should have picked up on your 'act of governing' answered by mollymolloy's 'process of governing', but the retention of the word government got in the way. What I didn't realise then is that I make a similar semantic distinction to yours — but in a narrower range of contexts when no human administration is involved.

    I'm delighted that I slowly reached the same conclusion as you two (and Mallamb, I think) without first appreciating the joke.

    ReplyDelete
  37. Correction

    but the retention of the word government got in the way

    What I should have said is the use of a word still containing govern got in the way.

    ReplyDelete
  38. I think I might know what's going on with Canepari's downward arrow: in “standard” and central/southern Italian, syllable-final nasals (which I'll transcribe as //n//) always share the point-of-articulation of the following consonant ([m] before bilabials, [ɱ] before labiodentals, [n] before coronals etc. -- all but the first transcribed as the phoneme /n/, probably due to the spelling which has «nf» for [ɱf] etc.), but I've read that there are northern dialects for which syllable-final nasals are always prevelar, no matter what follows. (I've never noticed this in “real life”, but I don't listen to that many northerners either.) So maybe some speakers use standard [mm] in careful speech but regional [ŋ̟m] colloquially, and [ŋ̟] would be considered to be /n/. (As for me, I think that I sometimes do have a few milliseconds of alveolar contact in San Marino in slow speech -- though almost completely overlapping with the labial closure, but this doesn't happen with most other sequences of //n// + bilabial, and I suspect that might have to do with the relatively low frequency of San Marino.)

    ReplyDelete
  39. Not a place that Italians have much need to think about, eh?

    ReplyDelete
  40. John's blog followers may want to read Jack Windsor Lewis's reply to be found here: http://www.yek.me.uk/Blog.html#blog368

    ReplyDelete
  41. Thanks for the link, Kraut.

    JWL's defence of his own listing of ɡʌvm̩ənt alone seems unassailable to an old TEFLer like me. His attack on John's listing would carry more weight if he could demonstrate that the other two gʌvᵊnmənt and gʌvᵊmmənt do not occur with any significant frequency. Merely showing that they are much less frequent that gʌvᵊmənt won't do.

    From a TEFL point of view, the only problem with John's listing is that the most frequently heard pronunciation (according to JWL's observations) is listed third. But that order allows consistent explanation in a minimum of characters. A more reasonable request might be that John should append some frequency information.

    JWL refers back to an earlier discussion where he compares John's treatment of attainment, entertainment, imprisonment. But surely there's a difference: these words have (for me) two variants -nm- and -mm-; government (we all seems to agree) has three variants -nm-, -mm- and -m-. (These are, of course somewhat idealised — not aspiring to the phonetic nuance of Mallamb's transcriptions.)

    It would be nice to see some statistics for the relative frequency of the various British pronunciations (and American?) — broken down into the meanings of 'regime' and 'process/fact of governing'.

    Afterthought: is there a similar dissociation between 'result' and 'process' with attainment and entertainment? I don't think I make a distinction but...

    ReplyDelete